A major educational goal is to build a knowledge base. To foster this goal, educators employ numerous techniques & strategies, including direct instruction, reading, & discussion, to name a few. Importantly, for education to have its maximum impact, learners must go beyond what is directly given to generate novel understandings. Without this capacity, learning would proceed exceedingly slowly, as each fact is acquired in turn. Fortunately, learners are not restricted to acquisition of information in this tedious way. Instead, they employ (deliberately or otherwise) a number of logical processes that permit valid extensions of knowledge through self-generation of new information. One such productive process is the subject of the proposed research, namely, self-derivation of new factual knowledge through integration of separate yet related episodes of new learning. The process requires that information newly learned in one episode (e.g., liquids expand with heat) be integrated with information newly learned in another (e.g., thermometers contain liquid) to derive new knowledge (thermometers work because liquid expands as heat increases). Such novel understandings are the currency of academic as well as vocational success. Our research has revealed both individual and age-related variability in this productive process in childhood. The mechanisms involved in this form of productive knowledge extension, & that contribute to age-related variability therein, have yet to be explicated. Moreover, whether individual variability is stable over time, and the domain-general cognitive abilities and component processes that relate to it, have yet to be examined systematically. Answers to these questions are vital because self-derivation of new knowledge through integration is linked to academic success in children (Esposito & Bauer 2017) & adults (Varga, Esposito & Bauer 2018). Accordingly, Aim 1 is to explicate patterns of age-related change and individual developmental trajectories in self-derivation and retention of new factual knowledge through integration across the elementary school years. This aim will be addressed in two cohorts of children tested at yearly intervals for 3 years: Cohort 1: ages 7, 8, & 9; Cohort 2: ages 9, 10, & 11. Aim 2 is to identify sources of variability in self-derivation and retention of new factual knowledge. We will test concurrent & longitudinal predictors from two categories: (A) General cognitive abilities including working & episodic memory (a.k.a. ?fluid? abilities) & the depth and breadth of accumulated semantic knowledge (a.k.a. ?crystallized? abilities; Aim 2a). (B) How effectively the component processes involved in integration & self-derivation are carried out (Aim 2b). By explicating the mechanisms of this productive process and the variables that limit its rate of development, we advance the ultimate goal of developing means of facilitating accrual of knowledge, and thus positive developmental, educational, and vocational outcomes.